Using Python for Digitizing

[TonySalomone]

Hi, I am wondering if it is possible to utilize python to automate a large digitizing task.  I can provide more details to whomever responds but basically I am collecting several linear datasets (bicycle and pedestrian facilities) that need to be consolidated into a statewide bicycle/pedestrian geodatabase.  I intend to apply these facilities to the statewide street centerline file so as to avoid heads up digitizing and so I can ensure topological accuracy.  BUT, I am hoping a python script will enable for automation of this task.  Any ideas are welcome.  Thanks!

Tony

[ChristopherThompson]

i'm honestly not sure how you might accomplish this using python but a couple of thoughts come to mind:

Are you able to do this in the context of a topology, if so, then editing the topology with appropriate rules can help manually editing these data be less painful

If these already follow existing routes, can you simply utilize those data rather than creating new polylines? I expect you would have to have a way to identify start/stop points to define the routes.  This process might be scriptable depending on how you go about it.

Without more details about the nature of the data you're collecting its hard to comment more.

[TonySalomone]

Hi!  Thanks for getting back to me.  I am using ArcView so I don't have access to the topology tools (bummer).  Here is the nature of the data:

Bike/ped facility data (bike lanes, sidewalks, wide shoulders, etc) are currently not snapped to any centerline shapefile. 
The data is coming from several jurisdictions within South Carolina. 
I am using South Carolina's statewide street centerline shapefile as the basis for the creation of a statewide bike/ped facility inventory that is to be current and topologically accurate.  The bike/ped shapefiles I have are offset by varying distance to this centerline shapefile.
Start/stops for each bike/ped facility do not necessarily connect from node to node. 

Thanks again and wish me luck!

Tony

i'm honestly not sure how you might accomplish this using python but a couple of thoughts come to mind:

Are you able to do this in the context of a topology, if so, then editing the topology with appropriate rules can help manually editing these data be less painful

If these already follow existing routes, can you simply utilize those data rather than creating new polylines? I expect you would have to have a way to identify start/stop points to define the routes.  This process might be scriptable depending on how you go about it.

Without more details about the nature of the data you're collecting its hard to comment more.

[ChristopherThompson]

I feel your pain!

The bike/ped shapefiles I have are offset by varying distance to this centerline shapefile.
Start/stops for each bike/ped facility do not necessarily connect from node to node.

This sounds very difficult, this would be a good use of an event table but if you don't have any reference to where along a route a particular trail or bikelane stops or starts that won't be of much help.  Scripting may help you to select existing routes that fall within a search distance or which share a name or something like that, at least that would help in whittling down the large number of polylines in your streets centerline data.  But from there, this sounds like a manual inspection and editing exercise.  Have you looked at the attribute transfer capabilities in the georeferencing toolbar? I could see some utility there for you perhaps.
In general what i'd do would be something like the following steps:

• make a duplicate of the streets centerline data


• weed that down to just those that match your trails/bike route data


• cut those polylines to match the endpoints of your trailes/bike routes and discard the irrelevant portions


• apply any offsets to segments


• use the attribute transfer tool to move your attributes from the trails/bike route data to the edited copy of the centerline data

[TonySalomone]

I like it.  I think there's still the huge task of manually spatially transferring attributes with the links tool, no?  And my attributes will all have to have the same naming conventions, correct?  I haven't used this tool since school...long ago. 
Thanks very much!

Tony

[ChristopherThompson]

And my attributes will all have to have the same naming conventions, correct?

Its been a little while since i used that tool also, but there may be a field mapping exercise you go through.  Best practices would suggest that you'd 'normalize' the tables so that there are similarly named/typed fields and that the value ranges are consistent.  The former could be done fairly simply by creating a new feature class with the structure you want in the final dataset.. and then using the Append tool with the 'no test' argument to force field mapping when you execute the append.  The latter.. making sure that values are consistent across data sets.. thats almost always a bigger issue.  But yeah, in the end I doubt there is an 'Easy Button' for this.

[PaulsonShelley]

Hey did you completed doing your work using python or still work is going on ?
discount fixie bikes